Why pre-screen counsellors in this area before the interview
This work attracts strong opinions about technology, and a counsellor who arrives with a position tends to argue with the client rather than work with them. What clients bring is ordinary: loneliness, avoidance, conflict with a partner about how much time goes to a system. That needs clinical training and a referral threshold, not a view on AI. A short screen tests both, starting with when they would hand a client on.
What actually matters when screening Human-AI Relationship Counsellor candidates
- 01
Clinical competence
Check their counselling training and licence (LPC, LMFT, LCSW, BACP) plus fluency with companion platforms such as Replika, Character.AI or custom GPT partners in casework.
- 02
Patient safety and protocol
Probe screening for suicidality, psychosis and AI-reinforced delusion, plus protocols when a chatbot has echoed self-harm ideation or a client resists platform separation.
- 03
Patient communication
Assess how they discuss an AI partner without ridicule or collusion: validating grief after a model deprecation, or a spouse discovering companion app use.
- 04
Working in a clinical team
Look for coordination with psychiatrists, couples therapists, trust and safety teams at AI vendors, and supervision or peer consultation on novel presentations.
Pre-screening questions to ask Human-AI Relationship Counsellor candidates
12 questions grouped by what they test. Ask the same set in every screen and score answers on a consistent scale, or send them as an async video screen and compare answers side by side.
Clinical grounding
3 questions01What training and clinical background do you bring to counselling in this area?
Listen forRecognised counselling training with supervision arrangements, plus specific reading in this area.
Interest in technology offered in place of clinical training, or no supervision arrangement.
02Can you describe a client conflict involving an AI system and how you worked through it?
Listen forThe client's underlying need identified, with the technology treated as the presenting issue rather than the problem.
The technology treated as the problem to remove, or the client's account argued with.
03Can you give an example of AI use improving a client's relationship with another person?
Listen forA balanced view that recognises benefit as well as harm, drawn from actual client work.
A settled position either way, or no case where the technology helped a client.
Attachment handled
3 questions04How do you work with a client who has formed a strong emotional attachment to an AI system?
Listen forThe attachment explored without judgement, with the function it serves for the client understood first.
Attachment treated as pathology to correct, or the client told the relationship is not real.
05How do you help clients set boundaries with AI tools in their daily life?
Listen forBoundaries the client sets themselves, tied to what they said they wanted rather than imposed.
Usage limits prescribed by the counsellor, or abstinence recommended as a default.
06What is your experience with clients using AI emotional support between sessions?
Listen forAwareness of what these tools do badly, with the client's between-session use discussed openly.
Between-session tool use dismissed, or its risks in a crisis not understood.
Knowing when to refer
3 questions07How do you work with a client who feels dismissed or misunderstood by an AI system?
Listen forThe feeling taken seriously and linked to the client's wider pattern rather than explained away technically.
The client corrected about how the system works, or the feeling treated as a misunderstanding.
08How would you support a client whose AI assistant acted against what they asked for?
Listen forPractical support alongside the emotional work, with escalation if the client is at risk.
No risk assessment mentioned, or distress at this level treated as purely technical.
09How do you help clients cope when an AI system's behaviour changes unpredictably?
Listen forLoss and adjustment recognised as real, with a clear threshold for referring a client on.
No referral threshold named, or every presentation treated as within their scope.
Consent and privacy
3 questions10What ethical considerations do you apply when counselling in this area?
Listen forProfessional ethics applied as normal, with awareness of where this area lacks settled guidance.
Ethics answered as opinions about AI, or professional codes not referenced at all.
11How do you think about consent when a client's AI use involves other people?
Listen forThird parties considered, particularly where a client's system holds another person's information.
Only the client considered, or no thought given to who else is affected.
12How do you address client privacy concerns about their AI interactions?
Listen forHonest about what these services retain, with session confidentiality kept separate from the client's own tools.
Privacy reassurance offered with no knowledge of retention, or client material entered into a third-party tool.
How to score responses
Score every candidate on the same four criteria immediately after the screen. At this stage you are shortlisting for panel interviews, not making the final call.
Clinical competence
35%5Holds a current counselling licence, cites modalities used (CBT, EFT, ACT) and describes real clients presenting with AI attachment.
Patient safety and protocol
30%5Names a structured risk protocol, describes escalation to psychiatry, and distinguishes healthy companion use from dependency with clear thresholds.
Patient communication
20%5Uses the client's own language for the AI, holds neutrality, and gives a concrete example of de-escalating shame in session.
Working in a clinical team
15%5Brings cases to supervision, documents referrals cleanly, and has worked alongside platform safety staff or clinical researchers on attachment cases.
Clients bring loneliness and conflict, not a debate about technology. A one-way video screen asks when they would refer a client on.
Try it on HirevireScreening FAQ
Process basics
How long should a pre-screening round for this role take?
Fifteen minutes across eight to ten questions, answered async. Enough to establish clinical grounding, test how they handle attachment, and check their referral threshold and confidentiality practice.
What should I verify outside the screen?
Registration, supervision arrangements and insurance, in full. This is client-facing work with vulnerable people, and no screen substitutes for confirming someone is licensed to practise where you operate.
Evaluating answers
What is the strongest signal when screening this role?
A clear referral threshold. Counsellors with clinical grounding name what takes a client beyond their scope and to whom. Anyone who would work with anything presented has not been trained safely.
What should worry me in an answer?
A settled position on whether AI relationships are good or bad. Both positions get argued at the client. What you want is curiosity about what the client is getting from it and what it costs them.
Go deeper on this role
Sanat has been hiring since 2012 and watching the recruitment industry change up close ever since, and turned that screening process into Hirevire's video screening platform. LinkedIn
